What is Poco Claw used for?

Poco Claw is a web-based AI coworker built on a Claude Code agent, with channels, sandboxed task execution, MCP and skills support, positioned as an OpenClaw alternative. It is a good fit for delegating coding tasks to a shared AI agent, running agents safely away from the host machine and reviewing what an agent did through playback.

Is Poco Claw open source?

Yes. Poco Claw is open source under the MIT licence. Its source code is on GitHub at poco-ai/poco-claw and is written mainly in Python.

Yes. Poco Claw is open source, so the software itself is free to use.

Can I self-host Poco Claw?

Yes. Poco Claw can be self-hosted on your own server or infrastructure.

What is Poco Claw an alternative to?

Poco Claw is an open-source alternative to Manus. Other open-source alternatives to Manus include LobsterAI, LemonAI and AutoGPT.

Is Poco Claw actively maintained?

Yes. The most recent commit to Poco Claw was on 22 September 2026, and the latest release is 0.5.7, published on 18 June 2026. The project has 1.4k stars on GitHub.

Poco Claw: open-source alternative to Manus

About Poco Claw

Poco, or Poco Claw, is an open-source AI agent workspace powered by a Claude Code-based agent. It is presented as a more polished and safer alternative to OpenClaw, with a web interface, built-in messaging support and a sandboxed runtime that keeps tasks isolated from the host machine.

Teams work in long-lived servers with channels, direct messages and reply threads, can mention persistent agents, turn messages into trackable tasks and publish outputs to shared files. Every task runs in an isolated container, with optional mounting of local directories when self-hosting. Other features include plan mode, conversation queueing, project settings for model and Git repository, an artifacts view for HTML, PDF, Markdown and diagrams, a playback view for commands and browser sessions, MCP and skills import, a built-in browser and scheduled background runs.

The project is written in Python, licensed under MIT, and has documentation hosted separately. It can be self-hosted, and the README notes that local directory mounting is available only in that mode. It suits developers and small teams who want to delegate coding and research work to a shared, observable agent.

Key features

Channels, DMs, and threads with agents
Isolated container sandbox for tasks
Artifacts viewer for many file formats
Playback of commands and browser sessions
MCP and Skills import
Scheduled background execution

Good fit for

→Delegating coding tasks to a shared AI agent
→Running agents safely away from the host machine
→Reviewing what an agent did through playback

Built with: Python
Tags: ai-agent; claude-code; sandbox; mcp; chat; collaboration; self-hosted; openclaw-alternative; python

Poco Claw: questions and answers

What is Poco Claw used for?: Poco Claw is a web-based AI coworker built on a Claude Code agent, with channels, sandboxed task execution, MCP and skills support, positioned as an OpenClaw alternative. It is a good fit for delegating coding tasks to a shared AI agent, running agents safely away from the host machine and reviewing what an agent did through playback.
Is Poco Claw open source?: Yes. Poco Claw is open source under the MIT licence. Its source code is on GitHub at poco-ai/poco-claw and is written mainly in Python.
Is Poco Claw free?: Yes. Poco Claw is open source, so the software itself is free to use.
Can I self-host Poco Claw?: Yes. Poco Claw can be self-hosted on your own server or infrastructure.
What is Poco Claw an alternative to?: Poco Claw is an open-source alternative to Manus. Other open-source alternatives to Manus include LobsterAI, LemonAI and AutoGPT.
Is Poco Claw actively maintained?: Yes. The most recent commit to Poco Claw was on 22 September 2026, and the latest release is 0.5.7, published on 18 June 2026. The project has 1.4k stars on GitHub.